Skip to content

Mistral-Small-Instruct-2409

A 22B instruction tuned model for long inputs

Visit official site
Category AI models

What Mistral-Small-Instruct-2409 is

The model is described as a 22B parameter instruction tuned system for long text and complex instruction work.

What you can do with it

  • Handle a sequence up to 128k long
  • Call functions from a prompt

Getting started

  • Install vLLM and mistral_common
  • Load the model and set its parameters

Pros & cons

✓ What we like

  • vLLM is recommended for production inference
  • Server and client setups are supported

! What to watch out for

  • vLLM and mistral_common must be installed
  • Parameters such as token limit are set by the user

FAQ

How long can the input be?

The material states sequences up to 128k are supported.

What library is recommended?

vLLM is named for production level inference.

Last reviewed: 2026-11-15

More LLM API platform tools

View all →

How we review